2023 Volume 70 Issue 5 Pages 519-528
Prader-Willi syndrome (PWS) is a multisystem disorder with increased mortality predominantly due to obesity-associated complications; therefore, the management of obesity has been centric to therapeutic strategies for PWS. Although a multidisciplinary team approach has been successful for this purpose during childhood, it is generally difficult to implement during adulthood because of the lack of a structured transitional care program. A more detailed understanding of the current medical conditions of adults with PWS is needed to establish this program; however, limited information is currently available on this issue in Japan. Accordingly, we performed a questionnaire-based survey on 425 patients with PWS. There were 162 adult patients aged 18 years or older with median body mass indexes (kg/m2) of 29.4 and 30.4 in males and females, respectively. The frequencies of type 2 diabetes mellitus (T2DM) and hypertension in adults with PWS were 40.4 and 19.4%, respectively. Growth hormone (GH) therapy during childhood correlated with lower rates of T2DM and hypertension during adulthood. Among adult patients, 54% were treated by pediatricians, whereas 44% were seen by internists with an endocrinologist/diabetologist being the most prevalent. Adult patients treated with GH during childhood showed a higher rate of being seen by pediatricians than those without, demonstrating that the multidisciplinary team approach, typically applied with GH therapy, may be continuously provided even after they reach adulthood. These results emphasize the importance of the seamless provision of the multidisciplinary team approach, which is of clinical importance for establishing an optimal transitional care program for PWS.
PRADER-WILLI SYNDROME (PWS; OMIM#176270) is a disorder that is caused by the lack of expression of paternal genes in the chromosome 15q11-13 region, mostly due to a deletion, maternal uniparental disomy (mUPD), or an imprinting defect [1-3]. The clinical features of PWS include characteristic facial features, mental and developmental delays, severe hypotonia with feeding difficulties during infancy, hyperphagia-associated obesity, and endocrine abnormalities, including growth hormone (GH) deficiency [1-3]. Since obesity-related comorbidities are associated with increased mortality in PWS [4-6], its management has been centric to therapeutic strategies for PWS. To achieve this, nutritional interventions and GH therapy, the latter of which effectively improves the body composition by decreasing fat mass and increasing lean mass, have been widely used with successful health outcomes [7-11].
The effectiveness of a multidisciplinary team approach including pediatric endocrinologists, dieticians, nurse practitioners, geneticists, and psychologists for the management of the body composition and obesity-related complications in children with PWS has been demonstrated [12-14]. Although the majority of children with PWS are treated with GH, which allows for the provision of frequent and sustained interventions by the multidisciplinary team, the multidisciplinary team approach is generally difficult to implement during adulthood, which is partly due to the lack of a structured transitional care program for PWS [12, 15]. Transitional care is a process during which dedicated health care programs are provided when care is transferred from pediatric to adult caregivers [16]; however, transitional care in PWS has been difficult and challenging because of the multisystem nature of this disorder, which includes cognitive and psychiatric symptoms [15, 17].
A more detailed understanding of the current medical state of adults with PWS is needed to provide a better transitional care program to patients with PWS. Although evidence on the health issues of adults with PWS overseas is increasing [18, 19], limited information is currently available in Japan. Based on these backgrounds, we herein performed a questionnaire-based survey to clarify the current medical conditions of adults with PWS in Japan.
The present study was approved by the Ethical Committee of Hamamatsu University School of Medicine (approval No. 18-119). Written informed consent was obtained from the patient and/or their guardian.
A questionnaire-based surveyA familial-response questionnaire was sent to 641 families who belonged to the patients’ association for PWS, named “Takenoko-no-kai”, in November 2018, and 425 (66.3%) responded by the end of January 2019. Takenoko-no-kai was founded in 1991 by family members of PWS patients with aims to improve the medical, educational, welfare and social environments surrounding PWS patients and there are 20 branches across the nation. Family members, including parents, were expected to answer the questionnaire. The questionnaire consisted of questions on background information that included age (year) and sex, medical information, and information on health care providers in cases aged 18 years or older. Medical information included height and weight, age at the diagnosis of PWS, the molecular basis for a genetic diagnosis, GH therapy, comorbidities including type 2 diabetes mellitus (T2DM), hypertension, sleep apnea, and scoliosis, and information on sex steroid replacement therapy (SSRT). Queries on comorbidities were performed by selecting a yes or no answer. There was also a question on age at diagnosis. A query for the molecular basis for a genetic diagnosis consisted of 2 questions, the first of which was intended to ask whether the diagnosis was based on a genetic analysis. The second question was applied to those diagnosed by a genetic analysis, which was performed by selecting the genotypes from the three categories of deletion, mUPD, and others. BMI was calculated by dividing body weight (kg) by height squared (m2).
Data analysisCategorical variables are presented as frequencies and percentages and a chi-square test was used for group comparisons. Other variables are expressed as medians (25% to 75% interquartile range: IQR) and the Mann-Whitney U test was utilized for group comparisons. A multiple regression analysis was also utilized to analyze differences with adjustments for age and sex as indicated in the text. A correlation analysis was performed using Spearman’s rank correlation coefficient. Statistical analyses were conducted using JMP software version 14 (SAS Institute, Cary, NC). P-values less than 0.05 were considered to be significant.
A total of 425 patients participated in the present study, and included 215 males and 199 females. Eleven subjects did not provide information on sex (Table 1). The median age of patients was 14 years (IQR: 5–23) and ranged between 0 and 48 years (Fig. 1A), with 162 subjects being equal to or older than 18 years, who were hereafter referred to as adult patients. Information on age was not provided in 6 subjects. The diagnosis of PWS was based on a genetic analysis in 397 subjects (93.4%) and the majority were diagnosed during the neonatal and infantile periods (Fig. 1B). The deletion type was the most prevalent (68.5%), followed by mUPD (23.3%) (Table 1). GH therapy was performed during childhood for 308 patients (72.8%) (Table 1). Two patients were treated with GH during adulthood; however, information on the underlying diagnosis for the use of GH was not provided. The percentages of patients with T2DM and hypertension were 17.0 and 7.7%, respectively, and age at diagnosis peaked at approximately 20 years (Fig. 2). A sub-analysis with patients older than 10 years old showed that the frequency of T2DM was 20.5%. There were four patients (3 males and 1 female) who developed T2DM at ages younger than 10 years (Table 2). Although BMI at the diagnosis of T2DM was not available, current BMI was lower than 25 in 2 male patients, whereas 1 male and 1 female patient showed increases in BMI to 27.2 and 34.1, respectively (Table 2). No patients had a history of GH therapy (Table 2). The percentage of patients with sleep apnea was 15.6% (Table 1) and age at diagnosis showed biphasic peaks at approximately 0 and 18 years old (Fig. 2). Scoliosis was reported in 31.7% of patients (Table 1) and a higher frequency was observed between 0 and 3 years and at 12 years (Fig. 2).
| Number N = 425 | Median (IQR) or frequency | Range | |
|---|---|---|---|
| Age, year | 419 | 14 (5–23) | 0–48 |
| M/F | 414 | 215/199 | |
| Genetic testing | 397 | 93.4% | |
| Deletion | 272 | 68.5% | |
| mUPD | 92 | 23.3% | |
| Others | 14 | 3.5% | |
| Non-deletion | 3 | 0.8% | |
| Unknown | 16 | 4.0% | |
| Age at diagnosis*, year | 391 | 0 (0–1) | 0–29 |
| GH therapy during childhood* | 308 | 72.8% | |
| T2DM* | 72 | 17.0% | |
| Hypertension* | 32 | 7.7% | |
| Sleep apnea* | 64 | 15.6% | |
| Scoliosis* | 134 | 31.7% |
*: Data were not available for 2 (GH therapy), 2 (T2DM), 9 (hypertension), 15 (sleep apnea), and 4 (scoliosis).
IQR: interquartile range
mUPD: maternal uniparental disomy
GH: growth hormone, T2DM: type 2 diabetes mellitus
Clinical characteristics of subjects.
A. The age distribution of subjects (N = 419). Information on age was not provided by 6 patients. B. Age at the genetic diagnosis is shown (N = 390). Among 397 patients with genetic testing, age at the diagnosis was not available for 7. N/A: information on sex is not available.
Age at the diagnosis of obesity-related complications.
Ages at the diagnosis of type 2 diabetes mellitus (N = 68), hypertension (N = 28), sleep apnea (N = 57), and scoliosis (N = 112) are shown. Ages at the diagnosis of type 2 diabetes mellitus, hypertension, sleep apnea, and scoliosis were not available in 4, 4, 7, and 22 patients, respectively. N/A: information on sex is not available.
| Sex | Age at onset (year) |
Current age (year) |
Current BMI (kg/m2) |
GH therapy during childhood | |
|---|---|---|---|---|---|
| 1 | M | 5 | 16 | 27.2 | None |
| 2 | M | 9 | 16 | 21.9 | None |
| 3 | M | 9 | 21 | 23.0 | None |
| 4 | F | 9 | 27 | 34.1 | None |
BMI: body mass index, GH: growth hormone
The characteristics of 162 adult patients (82 males, 78 females, and 2 unspecified) are shown in Table 3. Genetic testing was performed on 143 subjects, and 19 were clinically diagnosed. Median height in males was 159.8 cm (IQR: 154–164.2), which corresponded to –1.90 standard deviation (SD), whereas it was 147 cm (IQR: 141–150) in females, corresponding to –2.09 SD (Table 3). Median BMI in males and females were 29.4 (IQR: 23.7–38.1) and 30.4 (IQR: 24.9–34.2), respectively, and 67.1 and 74.7% of males and females, respectively, had BMI higher than 25 (Table 3). Age positively correlated with BMI in males (rs = 0.37, p < 0.001) and females (rs = 0.29, p = 0.011). The frequencies of patients with T2DM and hypertension were 40.4 and 19.4%, respectively, and no sex differences were observed (Table 4). The frequency of T2DM or hypertension did not significantly differ between those with deletions or mUPD (Table 4). The percentage of T2DM or hypertension was higher in those with BMI equal to or higher than 25 than in those with BMI lower than 25 (Table 4), suggesting that obesity is a risk factor for the development of T2DM and hypertension.
| N | Median (IQR) | Range | |
|---|---|---|---|
| Total | |||
| Age, year | 162 | 25 (22–32) | 18–48 |
| Genetic testing | 143 | ||
| Deletion | 104 | 24 (21–30) | 18–48 |
| mUPD | 17 | 22 (20–25.5) | 18–39 |
| Others | 10 | 26 (21.5–35.8) | 18–42 |
| Unknown | 12 | 28 (24.8–33.5) | 23–38 |
| Male* | |||
| Age, year | 82 | 24 (22–30) | 18–42 |
| Height, cm | 79 | 159.8 (154–164.2) | 130–174 |
| Weight, kg | 79 | 76 (57.6–93.5) | 30–160 |
| BMI, kg/m2 | 79 | 29.4 (23.7–38.1) | 15.4–67.5 |
| Deletion | 53 | 26.9 (22.8–38.0) | 15.4–64.1 |
| mUPD | 7 | 32.9 (27.0–38.8) | 21.1–48.5 |
| Female* | |||
| Age, year | 78 | 27 (22–34) | 18–48 |
| Height, cm | 75 | 147 (141–150) | 130.6–160 |
| Weight, kg | 75 | 63 (53.0–74.5) | 33–158 |
| BMI, kg/m2 | 75 | 30.4 (24.9–34.2) | 15.1–72.1 |
| Deletion | 45 | 29.3 (23.6–35.6) | 21.1–48.5 |
| mUPD | 10 | 29.5 (23.5–32.5) | 16.6–36.0 |
Information on sex was not available for 2 subjects.
*: Information on anthropometric parameters was not available for 3 subjects
IQR: interquartile range
mUPD: maternal uniparental disomy
BMI: body mass index
| N | Percentage | p-value | |
|---|---|---|---|
| T2DM (n = 161) | 65 | 40.4% | |
| M (n = 82) | 36 | 43.9% | |
| F (n = 78) | 29 | 37.2% | |
| M vs. F | 0.39 | ||
| BMI ≥25 (n = 111) | 52 | 46.9% | |
| BMI <25 (n = 44) | 9 | 20.5% | |
| BMI ≥25 vs. BMI <25 | 0.002 | ||
| GH+ (n = 92) | 28 | 30.4% | |
| GH– (n = 67) | 37 | 55.2% | |
| GH+ vs. GH– | 0.0017 | ||
| Deletion (n = 104) | 38 | 36.5% | |
| mUPD (n = 16) | 3 | 18.8% | |
| Deletion vs. mUPD | 0.16 | ||
| Hypertension (n = 155) | 30 | 19.4% | |
| M (n = 77) | 19 | 24.7% | |
| F (n = 76) | 11 | 14.5% | |
| M vs. F | 0.11 | ||
| BMI ≥25 (n = 105) | 25 | 23.8% | |
| BMI <25 (n = 45) | 3 | 6.7% | |
| BMI ≥25 vs. BMI <25 | 0.014 | ||
| GH+ (n = 90) | 8 | 8.9% | |
| GH– (n = 63) | 21 | 33.3% | |
| GH+ vs. GH– | 0.00015 | ||
| Deletion (n = 103) | 21 | 20.4% | |
| mUPD (n = 14) | 2 | 14.3% | |
| Deletion vs. mUPD | 0.59 |
Information on sex was not available for 1 and 2 subjects in T2DM and hypertension, respectively.
Information on BMI was not available for 6 and 5 subjects in T2DM and hypertension, respectively.
Information on GH therapy was not available for 2 subjects in T2DM and hypertension.
mUPD: maternal uniparental disomy
GH: growth hormone, T2DM: type 2 diabetes mellitus
BMI: body mass index (kg/m2)
Significant values are in bold.
The associations of GH therapy during childhood with BMI were investigated in adult patients. Two patients treated with GH during adulthood were excluded from the analysis. As shown in Table 5, among females, BMI was lower in patients treated with GH during childhood than in those without; however, this difference was not observed after adjustments for age. We also investigated the associations of GH therapy on the development of T2DM and hypertension and found that the percentage of these complications was significantly lower in those treated with GH during childhood than in those without (Table 4).
| GH-treated | Non-GH-treated | p-value | p-value | |
|---|---|---|---|---|
| Total | ||||
| N | 89 | 63 | ||
| Age, year | 23 (21–27) | 30 (25.5–36.5) | <0.001 | |
| BMI, kg/m2 | 28.0 (22.9–33.0) | 33.1 (26.6–39.9) | <0.001 | 0.065* |
| Male | ||||
| N | 51 | 27 | ||
| Age, year | 24 (22–27) | 26 (21–33.5) | 0.26 | |
| BMI, kg/m2 | 28.3 (23.2–36.2) | 34.2 (25.6–41.2) | 0.072 | 0.070** |
| Female | ||||
| N | 38 | 36 | ||
| Age, year | 22 (20–25) | 32.5 (28.5–38) | <0.001 | |
| BMI, kg/m2 | 27.9 (22.5–31.4) | 32.9 (28.6–39.8) | <0.001 | 0.12** |
*: adjustments for age and sex
**: adjustment for age
GH: growth hormone, BMI: body mass index
Significant values are in bold.
Information on SSRT was available for 156 adult patients (79 males and 77 females). It was administered to 21 males (26.6%) and 15 females (19.5%). Median ages at the initiation of SSRT in males and females were 18 years (IQR: 16–19.5, N = 21) and 15.5 years (IQR: 14.7–18.3, N = 14), respectively.
Health care providers for adult patients with PWSAmong 162 adult patients, information on health care providers was available in 160 (Table 6). One hundred and fifty-seven subjects (98.1%) were regularly seen by medical doctors. Eighty-seven patients (54.4%) were seen by pediatricians, whereas 70 (43.8%) were treated by internists. Eighteen patients were treated by both a pediatrician and an internist (11.3%), whereas 13.1% were not. Among internists, endocrinologists/diabetologists were the most prevalent. Medical doctors other than pediatricians/internists included psychiatrists/psychotherapists (38.1%), orthopedists (28.8%), and dermatologists (27.5%).
| Specialty | Number (N = 160) | Percentage (%) |
|---|---|---|
| Pediatrician | 87 | 54.4 |
| Internist | 70 | 43.8 |
| Endocrinologist/Diabetologist | 61 | 38.1 |
| Cardiologist | 16 | 10.0 |
| Pulmonologist | 11 | 6.9 |
| General internist | 5 | 3.1 |
| Both a Pediatrician & Internist | 18 | 11.3 |
| Not seen by a Pediatrician or Internist | 21 | 13.1 |
| Other specialists | ||
| Psychiatrist/Psychotherapist | 61 | 38.1 |
| Orthopedist | 46 | 28.8 |
| Dermatologist | 44 | 27.5 |
| Geneticist | 13 | 8.1 |
| Ophthalmologist | 10 | 6.3 |
| Otorhinolaryngologist | 9 | 5.6 |
| Gynecologist | 5 | 3.1 |
| None | 3 | 1.9 |
We investigated the relationship between GH therapy during childhood and health care providers during adulthood. The frequency of patients seen by pediatricians during adulthood was higher in those treated with GH during childhood (60 out of 95 subjects [63%]) than in those without (29 out of 65 subjects [45%]) (p = 0.002).
Medical conditions and the efficacy of transitional care in adults with PWS have been examined in detail overseas. van Abswoude et al. recently reported a median BMI of 33.3 (IQR: 26.8–41.4) in 354 adults with PWS with a median age of 31 years (IQR: 25–40 years) [20]. Comorbidities with a high prevalence include T2DM, sleep abnormalities, scoliosis, and a low bone mineral density [20-23], and mortality rates were shown to be higher in patients with PWS than in a reference population [24]. The efficacy of GH therapy during childhood to reduce mortality has been demonstrated. Proffitt et al. recently showed that deceased subjects received fewer interventions with GH during childhood than alive subjects [25]. Another study reported better mortality rates after than before the approval of GH to treat PWS in the US [26]. In addition to GH therapy, the efficacy of transitional care has been demonstrated. Paepegaey et al. showed that adults with PWS who received transitional care had lower BMI and better metabolic parameters [15]. The body of evidence on medical conditions in adults with PWS overseas is increasing, whereas limited information is currently available in Japan.
Accordingly, we herein investigated the current state of transitional care for adults with PWS in Japan and provided evidence to show that 54.4% of adults with PWS were followed by pediatricians, whereas 43.8% were treated by internists; however, the majority of internists were specialized medical doctors, with endocrinologists/diabetologists being the most prevalent, which is in accordance with the high prevalence of T2DM in adults with PWS. These results indicate that many adults with PWS are still being treated by pediatricians and that specialized internists, such as endocrinologists/diabetologists, are involved in the care of PWS when obesity-related comorbidities, such as T2DM, develop.
Obesity-related complications are the main cause of death in adults with PWS [4-6]; therefore, the management of obesity is centric to therapeutic strategies for patients with PWS throughout their lifetime. The present results revealed that median BMI in adults with PWS in Japan was approximately 30 and approximately 70% had BMI higher than 25; therefore, the majority of subjects were categorized as obese based on the criteria published by Japan Society for the Study of Obesity [27]. The prevalence of T2DM in all PWS patients was approximately 17% in the present study, which was consistent with previous studies [22, 25, 28, 29]. A recent study on 2,029 patients aged between 2 and 84 years reported a frequency of T2DM of 10.56% [25]. In Japan, Tsuchiya et al. showed that the frequency of T2DM in patients older than 10 years was 26.2% (17/65 patients). Oto et al. from the same group recently investigated the frequency of T2DM in 114 patients with PWS who were older than 10 years and showed that it was 28.9% [30]. Based on these findings, we performed an additional sub-analysis with patients older than 10 years and found a frequency of 20.5%. Although the age distribution of subjects in the present study differed from that in previous studies, the frequency of T2DM appears to have remained unchanged between 2011 and 2020 in Japan. It is important to note that there were 4 patients in the present study who were younger than 10 years when they developed T2DM. According to a review published in the US, the development of T2DM is rare during childhood [31]; therefore, the present results indicate that the early development of T2DM during childhood is unique to Japanese patients. Although information on BMI at the diagnosis of T2DM in these patients was not available, current BMI was not elevated in 2 male patients and mildly elevated in 1 male, whereas 1 female showed a marked increase in current BMI, suggesting a susceptibility to T2DM in Japanese patients with PWS, particularly in male patients, despite the lack of marked elevations in BMI. This notion is supported by evidence of east Asian males being susceptible to T2DM without elevations in BMI [32].
In the present study, we showed that GH therapy during childhood was associated with a lower frequency of T2DM and hypertension, which is consistent with previous findings [12]. Since GH therapy was approved in 2002 for PWS children with growth retardation in Japan, not all patients had a history of GH therapy during childhood. The mechanisms contributing to the relationship between GH therapy and the lower frequencies of T2DM and hypertension appear to be multifactorial; however, based on evidence for the beneficial effects of GH therapy on the body composition, namely, decreases in fat mass and increases in lean body mass [10], a better body composition may at least be partly responsible. However, there is also evidence to show that the body composition deteriorates after the cessation of GH therapy during adulthood [33-35]; therefore, there may be additional mechanisms contributing to the associations of GH therapy during childhood with the frequencies of comorbidities, which may be the provision of multidisciplinary team approach because it is more available when accompanied by GH therapy. The multidisciplinary team approach allows guardians to acquire more information on the management of PWS, including nutritional interventions, which will be maintained even after GH therapy is terminated. Additionally, transitional care may be more accessible to these individuals. It is important to note that PWS patients treated with GH during childhood were more likely to be followed by pediatricians during adulthood, which implies that the multidisciplinary team approach is continuously provided to these subjects even after they reach adulthood. However, the present study revealed that 13.1% of adult patients were not followed by pediatricians or internists, indicating that the multidisciplinary team approach is not provided to all patients. These results emphasize the importance of establishing a transitional care program and advocating the efficacy of the multidisciplinary team approach for the management of PWS.
SSRT was administered to 26.6 and 19.5% of male and female adult patients, respectively, which is considered to be low based on the high prevalence of hypogonadism in PWS patients, particularly in males [21, 36-38]. The low rate of SSRT may be characteristic to Japan because SSRT was administered to the majority of adult PWS patients with hypogonadism according to publications overseas [15, 20]. This is in part because guardians expressed concerns regarding the deterioration of psychological and behavioral symptoms after the initiation of SSRT; however, there is currently no information on the unfavorable influence of SSRT on psychological and behavioral conditions in PWS. Multiple reviews have indicated that a gradual increase in SSRT is tolerated well by adolescents with PWS [2, 39]. Since impaired skeletal health, including low bone mineral density, is prevalent in PWS [20] and is mainly caused by inappropriate SSRT, SSRT needs to be considered where appropriate. Moreover, the introduction of SSRT is supported by findings showing that SSRT increased the lean body mass and reduced the fat mass in male patients with PWS [40].
There are several limitations in this study. Although the number of patients included was relatively large despite the rare nature of the disorder, a questionnaire-based analysis may have impaired the accuracies of data acquisition and the diagnosis of comorbidities. Because this study was conducted on a single family-support group, the results may not represent the medical conditions of PWS patients in Japan. Additionally, despite multiple obesity-related complications being prevalent in PWS, only T2DM, hypertension, and sleep apnea are included in the questionnaire, while cardiovascular complications, which are important complications related to obesity-related mortality, are not. Furthermore, other endocrine complications, such as osteoporosis and hypothyroidism, are not included. Medications for T2DM were not systemically evaluated in the present study. Due to the questionnaire-based nature of the present study, we were unable to include variables that are important for regulating BMI, including nutritional interventions as well as age at the initiation of GH therapy and its duration; therefore, the associations of GH therapy with BMI and comorbidities may be affected by the lack of these variables.
In conclusion, we herein provided evidence to show the current medical conditions in adults with PWS including transitional care. We also showed that GH therapy during childhood was associated with a lower frequency of obesity-related complications during adulthood. Although the underlying mechanisms remain unclear, we assume that the multidisciplinary team approach, which is generally applied to patients treated with GH, is continuously provided to these patients even after they enter adulthood. These results emphasize the importance of the seamless provision of the multidisciplinary team approach to patients with PWS, which is of clinical importance for establishing an optimal transitional care program for PWS.
We thank Yasuko Fujisawa (Department of Pediatrics, Hamamatsu University School of Medicine) for assistance obtaining ethical committee approval for this study. The present study was supported by a Grant-in-Aid from the Ministry of Health, Labour, and Welfare, Japan (Nanchisei Shikkan Kokufuku Kenkyu Jigyo [Grant number: 30080201, to M.K., K.M., N.M., H.I., Y.T., R.H., and T.O.]).
All authors designed the study and analyzed data. M.K. wrote the manuscript. All authors read and approved the submission of the manuscript.
The authors have no conflicts of interest to disclose.
